Risk factor analysis and risk assessment model development for Crohn's disease with metabolic dysfunction-associated fatty liver disease in a Chinese population
Background Metabolic dysfunction-associated fatty liver disease (MASLD) frequently occurs as a hepatic complication in patients diagnosed with Crohn's disease (CD), influenced by numerous risk determinants. This study aimed to explore the clinical characteristics associated with MASLD in CD patients and to construct a tailored risk assessment model. Methods Patients with CD who were hospitalized at our gastroenterology unit between January 2021 and March 2025 were enrolled. Patients were divided into two groups based on hepatic ultrasonography, as well as cardiovascular metabolic risk assessment: MASLD and non-MASLD. Clinical variables, laboratory results, as well as histories of medication intake, were obtained for both groups. Putative risk factors were first identified in univariate analyses, further adjusted by least absolute shrinkage and selection operator (LASSO) regression. Multivariate logistic regression was finally performed to establish the risk assessment model, from which a nomogram was created. Model performance was assessed by the concordance index (C-index), calibration curve, and decision curve analysis (DCA). Internal validation was performed by bootstrap resampling (1,000 replicates) to obtain optimism-corrected performance estimates, Sensitivity analyses were performed to address incorporation bias. Result A total of 187 patients with CD were enrolled, among whom the MASLD prevalence was 20.9%. Multivariate logistic regression identified independent risk factors of MASLD, including age, body mass index (BMI), high-density lipoprotein (HDL), lactate dehydrogenase (LDH), and fecal calprotectin (FC). The model showed good discrimination for ultrasound-defined MASLD (C-index 0.915). The calibration curve demonstrated close correspondence between predicted probabilities and observed outcomes. Bootstrap validation yielded an optimism-corrected C-index of 0.900 (estimated optimism, 0.015), with an optimism-corrected calibration slope of 0.854 and intercept of −0.106, while DCA suggested potential net benefit across a wide threshold range. In sensitivity analyses, the model excluding BMI and HDL achieved an area under the receiver operating characteristic curve (AUC) of 0.804. Conclusion A nomogram incorporating age, BMI, HDL, LDH and FC showed promising performance in identifying ultrasound-defined MASLD in hospitalized Chinese patients with CD. External and prospective validation is required before the model can be used to guide further hepatic evaluation.
Authors
- Keyu Jia
- Jun Shen (ORCID: https://orcid.org/0000-0001-7206-1847)
- Chen Ye (ORCID: https://orcid.org/0000-0002-0048-6974)
- Shiying Yuan
- Yuqi Qiao (ORCID: https://orcid.org/0000-0003-0182-9606)
- Xueyun Zhang
Institutions
- Shanghai Jiao Tong University (CN)
- Renji Hospital (CN)
Publication Details
- Journal
- Frontiers in Medicine
- Published
- 2026-11-18
- DOI
- https://doi.org/10.3389/fmed.2026.1982320
- Primary Topic
- Liver Disease Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 0.00